IMO, microservice has two practice levels. Level 1, single codebase with multiple entrances (programs). At this level, the application already scales in horizontal and functional, while still has the benefit of code & data sharing. Level 2, eliminate code & data sharing, use RPC or MQ for communication, split the project into multipile repositories. This level might be regarded as the "ture" microservices which is considered harmful according to the blog post. Generally speak, if level 1 could fit your business, there is no need to go for level 2. If it does need level 2, well, it is the complexity itself which leads to the architecture, there is no shortcut.
BTW, for level 1 businesses, I have a boilerplate for Node.js & TypeScript you may want to give it a try: https://github.com/bsdelf/barebone-js